Original article Arch Argent Pediatr 2024;122(1):e202303001

Prevalence of osteopenia of prematurity before and after

implementing an early strategy with the use of calcium
and phosphate
María P. Carrascal Gutiérreza , María C. Janisa, Pablo H. Brener Dika , María F. Gallettia ,
Gonzalo L. Mariania

ABSTRACT
Introduction. With the use of aggressive parenteral nutrition in very low birth weight infants, alterations
in calcium and phosphate metabolism were detected. In 2016, a prevention strategy was implemented
through calcium phosphate monitoring and early supplementation. Our objective was to study whether
this strategy reduces the prevalence of osteopenia and to identify associated risk factors.
Population and methods. Quasi-experiment comparing the prevalence of osteopenia between two groups:
one after implementing the calcium phosphate monitoring and supplementation strategy (01/01/2017–
12/31/2019) and another prior to such intervention (01/01/2013–12/31/2015).
Results. A total of 226 patients were included: 133 in the pre-intervention period and 93 in the
post-intervention period. The overall prevalence of osteopenia was 26.1% (95% CI: 20.5–32.3)
and it was reduced from 29.3% (95% CI: 21.7–37.8) in the pre-intervention period to 21.5%
(95% CI: 13.6–31.2) in the post-intervention period, with no statistical significance (p = 0.19). In the
multivariate analysis, the NEOCOSUR score for risk of death at birth, use of postnatal corticosteroids, and
the intervention period were independently associated with osteopenia. Being born after the intervention
reduced the probability of alkaline phosphatase > 500 IU/L by 71%, regardless of the other variables
included in the model.
Conclusion. Calcium phosphate monitoring and early supplementation is a protective factor against the
development of osteopenia in very low birth weight infants.

Keywords: osteopenia; preterm newborn infant; calcium; phosphate; very low birth weight infant.

doi: http://dx.doi.org/10.5546/aap.2023-03001.eng

To cite: Carrascal Gutiérrez MO, Janis MC, Brener Dik PH, Galletti MF, Mariani GL. Prevalence of osteopenia of prematurity before and after implementing
an early strategy with the use of calcium and phosphate. Arch Argent Pediatr 2024;122(1):e202303001.

a
Department of Pediatrics, Division of Neonatology, Hospital Italiano de Buenos Aires, City of Buenos Aires, Argentina.

Correspondence to María P. Carrascal Gutiérrez: maria.carrascal@hospitalitaliano.org.ar

Funding: None.

Conflict of interest: None.

Received: 1-19-2023
Accepted: 5-9-2023

This is an open access article under the Creative Commons Attribution–Noncommercial–Noderivatives license 4.0 International.
Attribution - Allows reusers to copy and distribute the material in any medium or format so long as attribution is given to the
creator. Noncommercial – Only noncommercial uses of the work are permitted. Noderivatives - No derivatives or adaptations
of the work are permitted.

1
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

INTRODUCTION METHODS
Advances in perinatal care have increased the Study type and design: before-and-after
survival of very low birth weight (VLBW) preterm study comparing the prevalence of OP between
infants, giving rise to the challenge of achieving 2 groups of VLBW infants who received early and
a sequelae-free hospital discharge.1,2 In preterm “aggressive” PN, differentiated by early Ca and P
infants (PTIs), calcium (Ca) and phosphate (P) monitoring and supplementation during their stay
reserves are scarce; after birth, it is difficult in the NICU. Infants born between 01/01/2013 and
to achieve an adequate intake due to several 12/31/2015 were included in the pre-intervention
reasons: a delay in reaching full enteral feeding, group and those born between 01/01/2017 and
insufficient mineral content in breast milk (BM), 12/31/2019, in the post-intervention group.
intestinal immaturity limiting absorption, and limited Study population: all VLBW infants were
mineral intake in parenteral nutrition (PN).3–7 These included. Patients referred from other health
situations favor the development of osteopenia of facilities, those with congenital malformations,
prematurity (OP).8,9 The prevalence of OP is 20% those who died in the first 4 weeks of life, and
to 30% in VLBW infants and 50% to 60% in infants those born in 2016 were excluded, as this was
with a birth weight < 1000 grams.5,10 Associated the time in which both strategies may have
risk factors include low birth weight (LBW); overlapped.
prolonged administration of PN; administration of Intervention: all included patients started
postnatal corticosteroids, xanthines, furosemide, receiving PN since their admission to the NICU;
and alterations in vitamin D metabolism.10–12 enteral feeding with BM was indicated as soon
Postnatal growth restriction continues to be as it was available. The hospital does not have a
common despite recommendations combining breast milk bank. In the pre-intervention period,
“aggressive” PN (macronutrient intake since the first initial PN was given without the addition of Ca and
day of life) with early administration of BM.13–17 When P, continuing with individualized PN. In the post-
these recommendations were implemented, a high intervention period, the initial PN contained 48 mg/
prevalence of hypophosphatemia, hypercalcemia, kg of Ca and 38 mg/kg of P (1:1 molar ratio). A
and hypokalemia was observed during the first permanent supply of PN bags was ensured.
week of life.18,19 This condition was described as The change was communicated to the health
refeeding syndrome,18–21 related to insufficient Ca care team in charge of these patients. Ca and
and P intake, and it has been suggested that it P monitoring was standardized and the
would favor the progress towards osteopenia of requirement for corrections was agreed upon. After
prematurity.22 a 1-year washout period, the new nutritional
In 2016, at the neonatal intensive care strategy was implemented as standard practice.
unit (NICU) of Hospital Italiano de Buenos Aires, a Primary outcome variable: OP was defined
modification was introduced in nutritional support as an alkaline phosphatase (AP) level > 500 IU/L
protocols: Ca and P intake was started from birth, at any time prior to discharge (requested as per
through standard PN, and serum values were clinical criteria).
monitored during the first 48 hours of life. By Study variables: time of first phosphatemia
optimizing mineral intake, refeeding syndrome value, lowest phosphatemia value and highest
would be reduced. 5,23,24 The hypothesis of this calcemia value, requirement for phosphate
study was that, by preventing calcium phosphate supplementation and rapid corrections of
disorder in the first 2 weeks of life, the prevalence phosphate, AP value.
of OP would also decrease. The primary objective Hypercalcemia was defined as an
of the study was to compare the prevalence ionized calcium value > 1.35 mmol/L, while
of OP between 2 groups of VLBW infants who hypophosphatemia was defined as a value
received early PN, differentiated by early Ca and < 4 mg/dL, with values < 2 mg/dL considered
P monitoring and supplementation. The secondary severe hypophosphatemia.
objectives were to compare the prevalence of Demographic variables and/or variables
early hypophosphatemia/hypercalcemia and the related to clinical course were collected,
requirement for intravenous (IV) correction of such as birth weight (BW); LBW; gestational
phosphate between both groups, and to identify age (GA); NEOCOSUR score (probability of
OP-associated risk factors. neonatal death, score from 0 to 1: where 1
i s t h e m a x i m u m p r o b a b i l i t y ) ; 25 s e p s i s ;

2
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

necrotizing enterocolitis (NEC) (Bell the univariate analysis (p < 0.05) and/or reported
staging 2/3); intraventricular hemorrhage (IVH); in the bibliography as associated with OP were
bronchopulmonary dysplasia (BPD) (oxygen included as adjustment variables. The model
requirement at 36 weeks); postnatal administration that best explained the results was selected on
of corticosteroids, xanthines, and furosemide; and the basis of post-estimation tests: pseudo R2,
duration of PN. goodness of fit, ROC curve for the model. The
crude and adjusted odds ratio (OR) with their 95%
Analysis of results confidence interval (CI) and p value were reported.
Measures of central tendency and dispersion Any p value < 0.05 was considered statistically
were described for quantitative variables based significant. The STATA 13 software was used for
on distribution. Results were reported for analysis.
dichotomous variables with frequency measures. A consecutive non-probabilistic convenience
Continuous variables were compared using the sampling was performed, without estimating the
t test or the Mann Whitney test based on their sample size and analyzing 100% of VLBW infants
distribution, while categorical variables were that met the inclusion criteria, born in each period.
compared using the χ² test or Fisher’s exact test The study was approved by the Ethics
based on assumptions. Committee for Research Protocols.
The null hypothesis of equality in the
prevalence of OP between both groups was RESULTS
assessed. A univariate analysis was done to A total of 267 VLBW infants were born during
compare the remaining variables between both the study periods. A total of 226 patients were
groups. To identify OP-associated risk factors, included (Figure 1), while 41 were excluded
a bivariate analysis was performed to compare because they did not meet the inclusion criteria.
2 groups according to whether or not they had Their mean GA was 29.6 ± 2.4 weeks and
presented the primary outcome. A multivariate their median BW was 1138 grams (945–1330).
logistic regression analysis was used to adjust Table 1 shows the demographic and baseline
for potential confounders. Variables arising from characteristics; the only difference shown is the

Figure 1. Flow chart of patients

Pre-intervention group (1/1/2013 - 12/31/2015) Post-intervention group (1/1/2017 - 12/31/2019)

154 patients Total number of patients: 267 113 patients

Excluded (41)=

17 patients Death in the first month* (31) 14 patients

+
4 patients Malformations (10) 6 patients

133 patients Final n: 226 patients 93 patients

* Causes of mortality in order of frequency: sepsis, enterocolitis or intestinal perforation, pulmonary hemorrhage, pulmonary
embolism, hydrothorax.

3
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

administration of postnatal corticosteroids. study population was 397 IU/L (306–517). The
The variables related to calcium and 75th percentile, which was estimated to assess
phosphate metabolism and clinical course the biological plausibility of the selected cut-off
are shown in Table 2. The samples for AP point, was 517 IU/L. The overall prevalence of
measurement were collected at a median of OP was 26.1% (95% CI: 20.5–32.3) with 99%
4 weeks with an interquartile range (IQR) of 3–6 (n = 58) of AP samples > 500 IU/L collected
(pre-intervention: median of 4 weeks and IQR of after 3 weeks of life, while it was 45.2% (95%
3–6; post-intervention: median of 5 weeks and CI: 33.5–57.3) in the subgroup of infants with a
IQR of 3–7). The median AP level for the overall birth weight < 1000 grams. In the pre-intervention

Table 1. Demographic and baseline variables associated with osteopenia by period

Variable Pre-intervention period (n = 133) Post-intervention period (n = 93) p value

GA (weeks), mean (SD) 29.7 (2.3) 29.6 (2.3) 0.88*

BW (grams), median (IQR) 1170 (960–1345) 1095 (920–1310) 0.15**
Low birth weight, n (%) 46 (34.6) 31 (33.3) 0.84***
Male sex, n (%) 68 (51) 43 (46) 0.47***
NEOCOSUR score, median (IQR) 0.13 (0.05–0.29) 0.11 (0.04–0.35) 0.93**
Postnatal GCs, n (%) 17 (12.8) 23 (24.7) 0.021***
Caffeine, n (%) 102 (76.7) 69 (74.2) 0.67***
Furosemide, n (%) 51 (38.3) 37 (39.8) 0.83***
Days of PN, median (IQR) 10 (8–14) 13 (8–19) 0.12**

GA: gestational age; BW: birth weight; GCs: glucocorticosteroids; PN: parenteral nutrition; SD: standard deviation;
IQR: interquartile range; n: number.
* t test.
** Mann-Whitney test.
*** χ² test.

Table 2. Variables associated with calcium and phosphate metabolism and clinical course by period
Variable Pre-intervention period Post-intervention period p value
n = 133 n = 93
Calcium and phosphate metabolism
AP > 500 IU/L, n (%) 39 (29.3) 20 (21.5) 0.19***
AP in IU/L, median (IQR) 385 (303–529) 404 (314–483) 0.89**
First P measurement in days of life, median (IQR) 2 (1–4) 1 (0–1) < 0.001**
Lower P in mg/dL, median (IQR) 2.8 (2–3.6) 3.3 (2.9–3.6) < 0.001**
Higher iCa in mMol/mL, median (IQR) 1.43 (1.37–1.53) 1.47 (1.41–1.54) 0.051**
Hypercalcemia iCa > 1.35 mMol/mL, n (%) 109 (82) 90 (96.7) < 0.001
Hypophosphatemia < 4 mg/dL, n (%) 107 (80.4) 83 (89.2) 0.08***
Severe hypophosphatemia < 2 mg/dL, n (%) 27 (20.3) 1 (1.08) < 0.001****
Rapid correction of P, n (%) 19 (14.3) 0 (0) < 0.001****
P supplementation, n (%) 92 (69) 67 (72) 0.64***

Morbidities associated with clinical course

BPD, n (%) 30 (22.5) 29 (31.1) 0.14***
IVH, n (%) 21 (15.8) 18 (19.3) 0.48***
Sepsis, n (%) 23 (17.3) 23 (24.7) 0.17***
NEC, n (%) 5 (3.8) 12 (12.9) 0.02

**** P: phosphate; iCa: ionized calcium; AP: alkaline phosphatase; BPD: bronchopulmonary dysplasia;
IVH: intraventricular hemorrhage; NEC: necrotizing enterocolitis; IQR: interquartile range; n: number.
* t test.
** Mann-Whitney test.
*** χ² test.
**** Fisher’s test.

4
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

period, it was 29.3% (95% CI: 21.7–37.8); and it Lastly, a multivariate logistic regression
reduced to 21.5% (95% CI: 13.6–31.2) in the post- analysis was performed (Table 4). The baseline
intervention period. In the subgroup of infants risk of death (NEOCOSUR score), receiving
with a birth weight < 1000 grams, it decreased postnatal corticosteroids, and the period during
from 52.6% (95% CI: 36.5–68.1) to 37.8% which the birth occurred were independently
(95% CI: 23.5–54.7). These differences were associated with OP. Being born during the post-
not statistically significant for both strata. The intervention period reduced the probability of
minimum phosphatemia level, the prevalence of AP > 500 IU/L by 71%, regardless of the other
severe hypophosphatemia, and the requirement variables included in the model. Figure 2 shows
for rapid P corrections showed significant the corresponding ROC curve.
differences, just like the difference in days to
the first phosphatemia measurement, favoring DISCUSSION
the post-intervention group. Regarding neonatal This study shows that, for VLBW infants
morbidities related to prematurity, there were no receiving “aggressive” PN, the implementation of
differences between groups in terms of BPD, IVH, supplementation and close monitoring of Ca and
and sepsis. A significant increase in NEC was P is a protective factor against the development of
observed in the post-intervention period. OP. This is a common complication in the studied
Table 3 shows the results of the univariate population with variable prevalence according
analysis according to the presence or absence to their GA and BW, with those born before
of OP. Patients with OP were smaller and more 28 weeks of GA and with a BW < 1000 grams
immature and had a higher baseline risk of death. being the most affected ones. 9,12,20 These
The group of patients with OP received additional results are consistent with those reported in the
corticosteroids, diuretics, and PN for a longer bibliography, both in the overall population and
period and had a higher incidence of late-onset among PTIs with a birth weight < 1000 grams.
sepsis and NEC. The values corresponding to Although such reduction was not statistically
calcemia, phosphatemia, and the requirement significant after the intervention, it is possible
for rapid P corrections were statistically different. that the sample size was underpowered. In

Table 3. Univariate analysis by presence of primary result (alkaline phosphatase > 500 IU/L)
Variable/AP > 500 No (n = 167) Yes (n = 59) p value

GA weeks, median (IQR) 30 (29–32) 29 (27–30) < 0.001**

BW grams, median (IQR) 1205 (1005–1355) 940 (800–1125) < 0.001**
Low birth weight, n (%) 55 (33) 22 (37.3) 0.54***
Male sex, n (%) 78 (46.7) 33 (56) 0.22***
NEOCOSUR score, median (IQR) 0.08 (0.03–0.23) 0.36 (0.14–0.7) < 0.001**
Postnatal GCs, n (%) 17 (10.2) 23 (39) < 0.001***
Caffeine, n (%) 121 (72.5) 50 (84.7) 0.06***
Furosemide, n (%) 50 (30) 38 (64.4) < 0.001***
Days of PN, median (IQR) 10 (7–14) 16 (10–27) < 0.001**
Lower P in mg/dL, median (IQR) 3.2 (2.7–3.9) 2.7 (2–3.1) < 0.001**
Higher iCa in mMol/mL, median (IQR) 1.43 (1.38–1.53) 1.5 (1.41–1.57) 0.008**
Severe hypophosphatemia < 2 mg/dL, n (%) 15 (9) 13 (22) 0.009***
Rapid correction of P, n (%) 9 (5.4) 10 (17) 0.006***
Late-onset sepsis, n (%) 23 (13.7) 23 (39) < 0.001***
NEC, n (%) 8 (4.8) 9 (15.2) 0.018****
Post-intervention period, n (%) 73 (43.7) 20 (33.9) 0.19***

A: gestational age; BW: birth weight; GCs: glucocorticosteroids; PN: parenteral nutrition; NEC: necrotizing enterocolitis;
P: phosphate; iCa: ionized calcium; IQR: interquartile range; n: number.
* t test.
** Mann-Whitney test.
*** χ² test.
**** Fisher’s test.

5
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

Table 4. Univariate and multivariate models. Likelihood of alkaline phosphatase > 500 IU/L
Variable Univariate OR Univariate Multivariate aOR Multivariate
(95% CI) p value (95% CI) p value

NEOCOSUR score 56.9 (15–213) 0.001 46.3 (9–221) 0.001

Severe hypophosphatemia (P < 2 mg/dL) 2.8 (1.2–6.4) 0.01 0.9 (0.28–2.85) 0.87
PN days 1.06 (1.02–1.08) 0.001 1.03 (0.99–1.05) 0.10
Postnatal GCs 5.6 (2.7–11.6) 0.001 3.8 (1.3–10.8) 0.01
NEC 3.6 (1.3–9.7) 0.013 1.22 (0.28–5.3) 0.78
Post-intervention period 0.66 (0.35–1.22) 0.19 0.29 (0.12–0.7) 0.006

n = 225; pseudo R2 = 0.26; p = 0.001; goodness of fit p = 0.6 (Hosmer-Lemeshow test).

Area under the ROC curve of the model = 0.82.
OR: odds ratio; aOR: adjusted odds ratio; PN: parenteral nutrition; GCs: glucocorticosteroids; NEC: necrotizing enterocolitis.

addition, the multivariate model showed that phosphate intake and the development of OP
being born in the post-intervention period was a is still a matter of debate. Some authors have
protective factor against OP, regardless of the reported the relationship between an insufficient
other variables included. These findings may have intake of these micronutrients during the first
a highly relevant clinical impact. 8 weeks of life and the development of OP
One aspect that is worth noting is the value among newborn infants with a GA < 30 weeks
defined as diagnostic of OP. Ninety percent of and extremely LBW.1 The presence of
blood AP is produced by bone. AP may reflect hypophosphatemia is one of the earliest markers
bone turnover and is used as a parameter of of altered calcium and phosphate metabolism
mineralization.26,27 Although some authors used that could be related to the development of OP
higher cut-off points (700–800 IU/L),28 a recent and may occur from early postnatal life. 12 In
study reported that a value of 500 IU/L has a this study, it was observed that an early mineral
100% sensitivity and an 80% specificity. 5 The intake was associated with lower chances of
75th percentile of AP was 517 IU/L, which gives hypophosphatemia with requirement of phosphate
biological plausibility to the selected cut-off point. correction, which may have contributed to the
The association between calcium and reduction in OP.

Figure 2. ROC curve for the multivariate model

Area under the ROC curve = 0.82.

6
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

Calcemia values were higher in the post- so their impact cannot be analyzed. Another
intervention group. This may be due to the early limitation may be the time at which the highest
administration of Ca in PN from the day of birth AP value was measured; some authors consider
and to the fact that the doses administered may that its measurement in the third week of life is
exceed the needs at this stage. a predictor value; others consider that the best
The baseline characteristics of both groups time would be between 4 and 8 weeks of life.22,28
were similar, except for the use of postnatal Due to the observational design of this study, it
corticosteroids, so they were comparable. was not possible to establish a specific time for
The deleterious effect of corticosteroids on sample collection. Such variability may result in
bone mineralization among PTIs is well known a collection bias, although 99% of samples in
and an independent association with OP our population with AP > 500 IU/L were collected
was established. 29–31 However, there was a after the third week of life. Given the observational
trend toward a decrease in OP in the post- nature of this study, it is not possible to warrant
intervention group, which might have been more the causality between both outcome measures.
pronounced without the difference in the practice However, some of Hill’s criteria for causation are
of postnatal corticosteroid administration. We met, making the observed association relevant.32
did not observe significant differences in terms Consistent with previous studies,5,23 we observed
of morbidity associated with prematurity, temporality and biological plausibility in the
except for a higher incidence of NEC in the different correlations in both periods and in a
post-intervention group. This complication robust multivariate model according to the post-
prolongs the days of PN and delays the estimation tools used.
progression of enteral feeding, leading to a Based on these findings, it may be concluded
state of increased risk of OP. Even so, the post- that optimizing the use of calcium and phosphate
intervention group showed a lower prevalence in VLBW infants receiving PN is independently
of OP, despite having a higher proportion of associated with a lower probability of OP. This
such risk factor. This supports the importance of practice may show benefits in their long-term
early Ca and P supplementation as a preventive clinical course. Nevertheless, prospective studies
intervention for OP. confirming these findings would be useful. n
In the univariate analysis, patients with
a diagnosis of OP were more immature and REFERENCES
smaller, in addition to having a higher risk of 1. Viswanathan S, Khasawneh W, McNelis K, Dykstra C,
et al. Metabolic bone disease: a continued challenge in
death according to their NEOCOSUR score. extremely low birth weight infants. JPEN J Parenter Enteral
They also showed a higher frequency of postnatal Nutr. 2014;38(8):982-90.
corticosteroid use, diuretic use, days of PN, late- 2. Stoll BJ, Hansen NI, Bell EF, Walsh MC, et al. Trends
onset sepsis, lower phosphatemia levels, and in Care Practices, Morbidity, and Mortality of Extremely
Preterm Neonates, 1993-2012. JAMA. 2015;314(10):1039-
higher calcemia levels. All these factors have 51.
been associated with the development of OP.10–12 3. Pieltain C, de Halleux V, Senterre T, Rigo J. Prematurity
In the multivariate analysis, it was observed that and Bone Health. World Rev Nutr Diet. 2013;106:181-8.
being born in the post-intervention group served 4. Karpen HE. Mineral Homeostasis and Effects on Bone
Mineralization in the Preterm Neonate. Clin Perinatol.
as a protective factor against OP. On the contrary, 2018;45(1):129-41.
postnatal corticosteroid use and baseline severity 5. Abdallah EAA, Said RN, Mosallam DS, Moawad EMI, et al.
estimation were observed to be independent risk Serial serum alkaline phosphatase as an early biomarker
factors. These results are especially encouraging for osteopenia of prematurity. Medicine (Baltimore).
2016;95(37):e4837.
considering that a change in behavior, such as 6. Rayannavar A, Calabria AC. Screening for Metabolic
Ca and P supplementation and monitoring from Bone Disease of prematurity. Semin Fetal Neonatal Med.
the first hours of life, may have an impact on the 2020;25(1):101086.
clinical course and development of comorbidity in 7. Patel P, Bhatia J. Total parenteral nutrition for the very
low birth weight infant. Semin Fetal Neonatal Med.
the long term. 2017;22(1):2-7.
This study has certain limitations. Certain 8. Stalnaker KA, Poskey GA. Osteopenia of Prematurity: Does
practices and clinical course outcomes associated Physical Activity Improve Bone Mineralization in Preterm
with the outcome measures could have been Infants? Neonatal Netw. 2016;35(2):95-104.
9. Montaner Ramón A, Fernández Espuelas C, Calmarza P,
modified in the study period, such as the use of Rite Gracia S, Oliván del Cacho MJ. Factores de riesgo
non-invasive ventilation, neurodevelopmental y marcadores bioquímicos de la enfermedad metabólica
care, etc. These variables were not collected, ósea del recién nacido prematuro. Rev Chil Pediatr.

7
 Original article / Arch Argent Pediatr 2024;122(1):e202303001

2017;88(4):487-94. 22. Bonsante F, Iacobelli S, Latorre G, Rigo J, et al. Initial

10. Alizadeh Taheri P, Sajjadian N, Beyrami B, Shariat M. amino acid intake influences phosphorus and calcium
Prophylactic effect of low dose vitamin D in osteopenia homeostasis in preterm infants – it is time to change the
of prematurity: a clinical trial study. Acta Med Iran. composition of the early parenteral nutrition. PLoS One.
2014;52(9):671-4. 2013;8(8):e72880.
11. Dokos C, Tsakalidis C, Tragiannidis A, Rallis D. Inside the 23. Mulla S, Stirling S, Cowey S, Close R, et al. Severe
“fragile” infant: pathophysiology, molecular background, hypercalcaemia and hypophosphataemia with an optimized
risk factors and investigation of neonatal osteopenia. Clin preterm parenteral nutrition formulation in two epochs of
Cases Miner Bone Metab. 2013;10(2):86-90. differing phosphate supplementation. Arch Dis Child Fetal
12. Faienza MF, D’Amato E, Natale MP, Grano M, et al. Neonatal Ed. 2017;102(5):F451-5.
Metabolic Bone Disease of Prematurity: Diagnosis and 24. Senterre T, Abu Zahirah I, Pieltain C, de Halleux V, Rigo
Management. Front Pediatr. 2019;7:143. J. Electrolyte and Mineral Homeostasis After Optimizing
13. Clark RH, Thomas P, Peabody J. Extrauterine growth Early Macronutrient Intakes in VLBW Infants on Parenteral
restriction remains a serious problem in prematurely born Nutrition. J Pediatr Gastroenterol Nutr. 2015;61(4):491-8.
neonates. Pediatrics. 2003;111(5 Pt 1):986-90. 25. Marshall G, Tapia JL, D’Apremont I, Grandi C, et al. A
14. Sakurai M, Itabashi K, Sato Y, Hibino S, Mizuno K. new score for predicting neonatal very low birth weight
Extrauterine growth restriction in preterm infants of mortality risk in the NEOCOSUR South American Network.
gestational age ≤32 weeks. Pediatr Int. 2008;50(1):70-5. J Perinatol. 2005;25(9):577-82.
15. ElHassan NO, Kaiser JR. Parenteral nutrition in the neonatal 26. Harrison CM, Johnson K, McKechnie E. Osteopenia of
intensive care unit. Neoreviews. 2011;12(3):e130-40. prematurity: a national survey and review of practice. Acta
16. Dinerstein A, Nieto RM, Solana CL, Perez GP, et al. Early Paediatr. 2008;97(4):407-13.
and aggressive nutritional strategy (parenteral and enteral) 27. Rustico SE, Calabria AC, Garber SJ. Metabolic bone disease
decreases postnatal growth failure in very low birth weight of prematurity. J Clin Transl Endocrinol. 2014;1(3):85-91.
infants. J Perinatol. 2006;26(7):436-42. 28. Hung YL, Chen PC, Jeng SF, Hsieh CJ, et al. Serial
17 Ibrahim HM, Jeroudi MA, Baier RJ, Dhanireddy R, Krouskop measurements of serum alkaline phosphatase for early
RW. Aggressive early total parental nutrition in low-birth- prediction of osteopaenia in preterm infants. J Paediatr
weight infants. J Perinatol. 2004;24(8):482-6. Child Health. 2011;47(3):134-9.
18. Cubillos Celis MP, Mena Nannig P. Hipofosfemia en recién 29. Doyle LW, Davis PG, Morley CJ, McPhee A, et al. Low-dose
nacidos prematuros: un trastorno bimodal. Rev Chil Pediatr. dexamethasone facilitates extubation among chronically
2018;89(1):10-7. ventilator-dependent infants: a multicenter, international,
19. Brener Dik PH, Galletti MF, Fernández Jonusas SA, randomized, controlled trial. Pediatrics. 2006;117(1):75-83.
Alonso G, et al. Early hypophosphatemia in preterm infants 30. Baud O, Maury L, Lebail F, Ramful D et al. Effect of early low-
receiving aggressive parenteral nutrition. J Perinatol. dose hydrocortisone on survival without bronchopulmonary
2015;35(9):712-5. dysplasia in extremely preterm infants (PREMILOC): a
20. Bustos Lozano G, Soriano-Ramos M, Pinilla Martín MT, double-blind, placebo-controlled, multicentre, randomized
Chumillas Calzada S, et al. Early hypophosphatemia in trial. Lancet. 2016;387(10030):1827-36.
high-risk preterm infants: efficacy and safety of sodium 31. Ukarapong S, Venkatarayappa SKB, Navarrete C, Berkovitz
glycerophosphate from first day on parenteral nutrition. G. Risk factors of metabolic bone disease of prematurity.
JPEN J Parenter Enteral Nutr. 2019;43(3):419-25. Early Hum Dev. 2017;112:29-34.
21. Pająk A, Królak-Olejnik B, Szafrańska A. Early 32. Bradford Hill A. Ambiente y enfermedad: ¿Asociación o
hypophosphatemia in very low birth weight preterm infants. causación? Rev Cub Salud Pública. 2008;34(2). [Accessed
Adv Clin Exp Med. 2018;27(6):841-7. on: May 29th, 2023]. Available at: http://dx.doi. org/10.1590/
s0864-34662008000200015